Assembly of the mitotic inter-microtubule bridge complex clathrin-TACC3-ch-TOG: a hybrid structural biology approach

During cell division, chromosomes - the structures containing DNA within cells - must be divided equally between the two new cells. This is achieved by the mitotic spindle, which uses fibres composed of microtubules to direct the movement of chromosomes inside the cell. Some of these fibres are reinforced by "bridges" to ensure they can withstand the increased stresses involved. The shortest and most common sort of bridge is made up of three proteins - called clathrin, TACC3 and ch-TOG. At present, we have only a rough idea how the three proteins interact and how, by coming together, they are able to bridge between microtubules.

When the interactions between the bridge-forming proteins are disrupted, they are no longer able to bind the microtubules, and as a result the mitotic spindles are significantly weaker. Cancer cells depend on robust mitotic spindles to enable their proliferation, resulting in tumour growth. Drugs currently used to treat cancer bind to microtubules and destabilize mitotic spindles, but also affect the microtubules in normal cells. We believe that targeting the interactions between these three proteins could produce cancer drugs with fewer side effects than current treatments. To do this, we need to understand the interactions between the three proteins in great detail.

We now propose to investigate the interactions between clathrin, TACC3 and ch-TOG using methods that will reveal how the three proteins come together at the level of their constituent atoms. In the short term, we will use this information to build up a picture of how each interaction contributes to bridge formation. In the longer term, this will help us to decide which of the interactions would make the best candidate for the development of new cancer drugs. Having models of the interaction in atomic detail will accelerate the process of designing these drugs.

Sister chromatids are connected symmetrically to two spindle poles through bundles of microtubules (MTs) called kinetochore-fibres (K-fibres). Built to withstand tension forces, the microtubules within K-fibres are more stable than other spindle MTs, in part because they are connected through proteinaceous bridges, such as the CTC complex comprising clathrin, TACC3 and ch-TOG. The beta-propeller of clathrin and the coiled-coil domain of TACC3 are both required for MT-binding, and the two proteins are brought together upon phosphorylation of an extended peptide region in TACC3 that binds to the ankle region of clathrin. Here, through structural, molecular and cell biology approaches, we aim to determine the structures of the two key interactions within the complex and to define the overall architecture of the core MT-binding module.

1) Interaction between clathrin and TACC3. The ankle domain of clathrin does not have a canonical phospho-binding motif. We will establish the structural basis of phosphorylated TACC3 recognition and measure the contribution of phosphorylation and other key residues to binding.

2) Interaction between ch-TOG and TACC3. The role of ch-TOG in the complex is unclear. We will investigate whether it is a dynamic component, measure its effect on the clathrin/TACC3 interaction and on the interaction with MTs. We will determine the structure of the complex to discover how ch-TOG recognises the TACC3 coiled-coil.

3) Assembly of the core MT-binding module. We will assemble the two complex structures together with other biophysical data into a model of the core complex. This will define the relative position and orientation of the domains within the complex and give insights into how the phospho-regulated interaction of TACC3 and clathrin brings together the coiled-coil and beta-propeller domains to form a MT binding module. The model will show how the position of ch-TOG in the complex is consistent with its role.

Research career development. Training the next generation of biomedical research scientists is an important component of our work that has lasting impact. Our aim is that the PDRAs who work on this project will be in a strong position to apply for Fellowships either directly afterwards, or after gaining further experience. This approach is highlighted by a former PDRA, Charlotte Dodson, who was first author on 3 papers from the Bayliss group, including one in Science Signaling, and has now been awarded a Research Fellowship from Imperial College to establish her own group.

Cancer genes. Two of the three proteins that we plan to study in this project (ch-TOG and TACC3) are overexpressed in cancer cells, but their contributions to cancer development and the underlying mechanisms are unclear. Our work will contribute to a deeper understanding of the functions of these proteins, leading to new insights into their roles in human disease.

CTC complex inhibitors. Anti-mitotic drugs such as vincristine and docetaxol are among the most commonly used treatments in the cancer clinic, and form part of the front-line treatments for childhood acute lymphoblastic leukaemia, the most commonly occurring form, and lung cancer. The path to discovering improved treatments for these diseases is long and challenging. It requires accurate models that generate ideas for therapeutic development, as well as reagents and assays to support drug discovery. Pharmaceutical companies or charitable/academic drug discovery groups will incorporate our work into drug discovery projects. This will be done through our existing network of collaborators or, with the help of the Enterprise and Business Development Office in Leicester, with new partners. As our track record on Aurora-A and NEK2 kinases shows, we will play a direct role in developing therapeutics. Development of protein-protein interaction (PPI) inhibitors is at the forefront of current drug discovery, and specific aspects of our work together form a 'package' that will contribute to the development of inhibitors targeting the CTC complex:
i) The recombinant CTC proteins we make, and the methods used to make them will be used in assays for developing PPIs.
ii) Binding constants for interactions and the assays used to generate them are useful for formulating strategies for targeting an interaction and in the screening/validation of inhibitory compounds.
iii) High-resolution structures of the target interface will be critical in the design and refinement of PPI inhibitors. The methods used to generate these structures will be used in the development of inhibitors, for example the NMR spectra and assignments that underpin the structure of the 4alpha domain could be used for NMR-based screening of chemical inhibitors.

Outreach - Local charities and patients. Just knowing that relevant research is being done on the disease is helpful to cancer patients and their supporters because it shows that scientists are taking an interest and that progress will be made in tackling the disease. We work with local hospitals, charities (such as HOPE) and patient groups, giving tours of our laboratories and presenting our work. We plan to involve the PDRAs working on this project directly in such events, giving them the opportunity to showcase their work to the public. We believe that outreach events play a crucial role in demonstrating to the public that the funds they raise, through taxation or charity, are being used to carry out world-class research for public good.

Outreach - Future scientists. Modern biomedical research is an inspirational topic for exciting children and young adults about science. Bayliss gives talks in schools about his own work, which have to date been focused on cancer. This project encompasses aspects of chemistry, biology and physics, and will form the basis for inspirational future talks on basic science by both PIs aimed at schoolchildren who are studying these subjects.